Plano, Texas · Senior Business Analyst · Salesforce · AWS · AI

Turning enterprise requirements into systems that ship, integrate cleanly, and prove their value.

Senior Business Analyst at HotelKey, working at the scope of a solution architect. Five-plus years across hospitality SaaS, automotive, and financial services, sitting where enterprise clients, product, and engineering meet: running discovery, designing the Salesforce, AWS, and API integrations that follow, testing them to certification, and increasingly wiring GenAI into the delivery itself.

See the work Download résumé ↓
01About

Analyst by title, architect by scope.

I'm a Senior Business Analyst at HotelKey, where I sit between enterprise hotel brands and the engineers who build for them. I run discovery, write the specs, design the Salesforce and API integrations that follow, and stay on the hook until they're certified and running in production. Seven Salesforce certifications, an AWS one, and a habit of writing the SQL that checks whether the system did what we said it would. Before this: financial forecasting at CG Infinity and a Salesforce re-architecture for Daimler AG at Infosys.

How I work

01Discovery is a contract.

Definitions get agreed in writing before build, so UAT tests the logic we signed, not opinions.

02Prove it landed.

A passing API call isn't a passing integration. I query the tables to confirm the data arrived correctly.

03Declarative first.

Flows and configuration before code. Solutions the next admin can read survive the one after that.

04Every engagement carries a number.

If the outcome can't be measured, the work isn't finished. The strip above is the receipt.

Currently

Building Trowel for Salesforce, an AI agent that audits a Salesforce org · studying for Salesforce AI Specialist · shipping one project a week through September.

02Experience
2023 to NowAug 2023 – Present

Senior Business Analyst

HotelKey Inc. · Carrollton, TX · promoted from Business Analyst · solution-architecture scope across CRS, Salesforce, and AWS
  • Primary analytical liaison and client-facing point of contact for 9+ enterprise hotel brands (Hilton, IHG, BWH Hotels) on a hospitality SaaS platform serving 4,000+ properties, leading discovery sessions, managing stakeholder expectations, and owning the requirements lifecycle end to end.
  • Translated client and business-stakeholder requirements into user stories, acceptance criteria, and functional specifications, documenting data-mapping rules and end-to-end business process flows including operational edge cases and system dependencies. Implementation effort down 30% per brand post go-live.
  • Architected a Salesforce Sales & Support platform using Einstein AI, automated Flows, Apex, and Lightning Web Components, with brand modeled as a data dimension so onboarding a new brand became configuration rather than a project. Delivered +20% lead conversion with First Contact Resolution (FCR) sustained above 70%.
  • Performed integration testing and release-readiness validation across 10+ mission-critical CRS APIs, building and executing Postman collections against Swagger/OpenAPI specifications, inspecting request and response payloads, and verifying system-to-system communication. Reduced API-related support tickets 30%.
  • Defined User Acceptance Testing (UAT) criteria and executed partner certification and regression cycles against CRS sandbox and staging environments, guiding enterprise clients through validation and documenting pass/fail results for sign-off.
  • Wrote SQL queries and joins to audit reservation and booking data across database tables, verifying payload accuracy and data integrity, investigating client-reported integration issues and data discrepancies, identifying root causes, and routing them to engineering.
  • Supported delivery of enterprise reservation data products on AWS (Kinesis, Glue, Athena) for near-real-time booking visibility, and contributed to performance validation of a DynamoDB Accelerator (DAX) rollout against contractual latency SLAs.
  • Monitored integration SLAs for uptime, response time, and error rates, escalating breaches with root cause narrowed down; partnered with Product Managers and engineering leads on backlog prioritization and sprint planning in an Agile/Scrum environment. GenAI-assisted reporting cut program review cycles 20%.

Tools Salesforce (Sales & Service Cloud, Apex, LWC, Flows, Einstein) · Postman · Swagger/OpenAPI · SQL · AWS Kinesis, Glue, Athena, DynamoDB · Jira · Confluence · Agile/Scrum

2023Jan – May 2023

Financial Business Analyst

CG Infinity · Plano, TX
  • Delivered GenAI-augmented market insights and financial trend analysis using Python, Alteryx, and LLM-assisted summarization for banking and financial-services stakeholders, driving an 8% profit increase and launching two strategic initiatives.
  • Built AI-assisted forecasting models in Power BI on Snowflake with advanced DAX measures and Power Query transformations, reducing manual analysis time 35% and surfacing executive insights 2× faster.
  • Engineered optimized Snowflake data pipelines with automated data-quality checks, improving query performance and retrieval speed 30% and enabling real-time financial reporting.
  • Developed predictive financial risk models using Python regression and scenario analysis, contributing to a 10% reduction in risk exposure and supporting evidence-based executive decision-making.
  • Gathered requirements from business stakeholders and translated them into data models, KPI definitions, and dashboard specifications reviewed and signed off before build.
  • Built Power Automate workflows for recurring operational tasks, including automated retrieval and storage of 200+ documents that had previously been handled by hand.

Tools Python · SQL · Snowflake · Power BI (DAX, Power Query) · Power Automate · Alteryx · Excel · LLM summarization

2019 to 2021Sep 2019 – Dec 2021

Systems Engineer · Salesforce Business Analyst

Infosys Limited · Bangalore, India · client: Daimler AG (Mercedes-Benz) · OneCRM for the European market
  • Led Salesforce re-architecture for Daimler AG (Mercedes-Benz) using Apex and Lightning Web Components (LWC) across the European automotive retail and wholesale programme, reducing manual data entry 20% and improving data accuracy 15%.
  • Built scalable LWC components and declarative Flows aligned to the Salesforce Well-Architected Framework, eliminating data duplication 25% and accelerating cross-team workflows.
  • Collaborated across five cross-functional teams (business, IICS, testing, rollout, support), reviewing and improving 80+ CRM-centric software designs and resolving 40+ client stories to keep production rollouts above a 95% success rate.
  • Developed and deployed 10+ workflows, Flows, and process builders, and redesigned the partner community site for 200+ distinct audience types.
  • Established CI/CD pipelines with GitHub Actions, Bitbucket, and Jenkins including automated test coverage, achieving 95% defect-free deployments and reducing post-release issues 30%.
  • Led full-cycle User Acceptance Testing (UAT), onboarding, and hypercare support across five cross-functional sub-teams, achieving 90% CSAT and cutting average issue resolution time 40% through a structured knowledge base and triage workflow.
  • Supported data migration and sandbox-to-production release management, resolving complex migration requirements between developer, testing, and staging environments.
  • Designed real-time Salesforce and Tableau dashboards surfacing KPIs to senior leadership, directly contributing to a 23% increase in Sales & Marketing revenue.

Tools Salesforce (Apex, LWC, Flows, Process Builder) · GitHub Actions · Bitbucket · Jenkins · Tableau · Jira · Confluence · Agile/Scrum

2019Jun – Aug 2019

Sports Analyst

Hudl · Mumbai, India
  • Built and maintained performance databases for high-school and youth teams, and produced detailed video breakdowns across basketball, soccer, volleyball, and American football.
  • Ran quality checks on final analysis reports, an early version of the same instinct that later became SQL data audits and written pass/fail certification.

Tools Video analysis · Database management · QA reporting

03Case studies

Three pieces of work, in detail.

01 · Salesforce

One Sales & Support platform for a thousand hotels

Nine-plus brands on a single org: brand as a data dimension, declarative-first automation, Einstein where it changed a decision.

+20% conversion · 70%+ FCR→ 02 · APIs & integration

Certifying CRS integrations before they ship

Spec-as-contract, Postman collections, written pass/fail certification, SQL audits, and SLA tracking across 10+ mission-critical APIs.

30% fewer API support tickets→ 03 · AWS & data

Near-real-time reservation data on AWS

Kinesis, Glue, and Athena for live booking visibility, plus validating a DynamoDB DAX rollout against the SLAs brands actually enforce.

Validated · signed off→
04Building

What I'm building.

An AI agent that reads a Salesforce org's metadata and produces the Well-Architected assessment consultancies bill five figures for. It's the answer I'm working out to a question I keep hitting at work: what does agentic AI look like inside a governed enterprise platform?

Personal project · Phase 1 of 5

Smart City Travel Planner

github.com/PakshalShah30/smart-city-travel-planner ↗

Give it a city, a starting point, and a time budget; it returns a routed day plan through real places, using real street-network travel times, that gets you back where you started before the time runs out. Picking a day out is not a sorting problem: every attraction has a location, a visit duration, an opening window, and a value to a particular person, and choosing the best subset and the best order is the Orienteering Problem, which is NP-hard.

It is also a deliberate rebuild of my own 2018 university project of the same name. That version specified three itinerary modes, route optimization, ratings, and opening-hours awareness, and shipped a two-column SQL lookup over four rows of placeholder data. This repository is the honest version of that specification, and it tracks the gap on purpose.

Next.js 16TypeScriptPostgres + PostGISDrizzleVitestOSM / OSRM
Build progressPhase 1 of 5
  • Phase 0: scaffold, seven-table data model, PostGIS geography columns, CI green
  • Phase 1: greedy insertion solver, geo primitives, opening-hours parsing, Mumbai seed data, planner UI wired to the solver
  • Phase 2: OpenStreetMap ingestion and OSRM travel-time matrices across two cities
  • Phase 3: 2-opt and Or-opt local search, then a constrained solver with time windows, each benchmarked against the last
  • Phases 4 and 5: map surface, accounts, share links, end-to-end tests, launch
Personal project · early, paused

Trowel for Salesforce

github.com/PakshalShah30/trowel-for-salesforce ↗

An AI agent that reads a Salesforce org's metadata and produces the kind of Well-Architected assessment consultancies bill five figures for: dead automation, conflicting triggers, permission sprawl, and a prioritized remediation roadmap. The design principle is that findings are detected deterministically by dependency-graph queries, and the language model explains and prioritizes them but never discovers them, because a hallucinated finding would destroy the product's whole premise.

Currently early and paused while the travel planner takes priority. The scaffold, retrieval pipeline, and a deliberately seeded test org are in place; the parser, dependency graph, and eval suite are not yet written. Listed here because the architecture argument is worth reading, not because it is finished.

PythonClaude APISalesforce Metadata APInetworkxSQLite
Build progressFoundations
  • Metadata retrieval pipeline against a live Developer Edition org
  • Test org seeded with deliberate tech debt: dead Flows, a trigger and Flow competing for one field, unused fields, an over-permissioned permission set
  • Parser and dependency graph, then the first deterministic detectors
  • Hybrid retrieval, golden-dataset eval gate, agent loop, Well-Architected report

Also shipped

Salesforce · developer tooling

FlowLint

Static analysis for Salesforce Flow metadata: five rules that catch missing fault paths, hardcoded record IDs, recursion risk, dead automation, and two Flows fighting over the same field. Exits non-zero so it can gate a pull request.

Python · ElementTree · pytest · GitHub Actions View repository ↗ APIs · integration testing

CRS Certification Harness

Partner certification for reservation APIs as one command: a mock CRS with deliberately planted defects, an OpenAPI contract, a Postman collection run by Newman, and an audit-ready pass/fail sign-off document generated from the results.

FastAPI · Postman/Newman · OpenAPI · pytest View repository ↗ Observability · SLA management

SLA Sentinel

Integration SLA monitoring as a system instead of a spreadsheet: scheduled probes into SQLite, rolling uptime and p95 latency against contractual thresholds, breach alerts with a cooldown, and a dashboard that shows which integration is failing and by how much.

FastAPI · SQLite · Chart.js · pytest View repository ↗
05Capabilities

What I bring to a room.

Salesforce platform7× certified

Apex, Lightning Web Components, Flows, Einstein AI, Data Cloud. Well-Architected by habit: declarative first, code where it earns its place, CI/CD around all of it.

AWS & cloud dataCertified Cloud Practitioner

Kinesis, Glue, Athena, DynamoDB, S3, Lambda, API Gateway. Reservation data pipelines and near-real-time booking visibility; validating API performance and SLAs on DynamoDB-backed services.

Integration & APIsREST · SOAP · OAuth

Postman collections, certification cycles, SLA monitoring, payload auditing with SQL. I treat an integration as untested until the client's data has landed and been queried.

Applied AILLM · RAG · agents

Claude, Einstein, prompt engineering, hybrid retrieval, LLM-as-judge evals. The interesting part isn't the model; it's the eval gate that makes it safe to ship.

Analysis & deliverySenior BA · solution design

Discovery, user stories, acceptance criteria, data mapping, UAT, partner certification, backlog prioritization, Agile/Scrum, Jira & Confluence. Client-facing by default; every requirement traceable to a test and a metric.

DataSQL · Snowflake · BI

SQL, Snowflake, dbt, Power BI, Tableau, Python, Alteryx. Pipelines with quality checks, dashboards leadership actually opens.

06Credentials

Verified, not self-declared.

SalesforceCertified Platform Developer II
SalesforceCertified Data Cloud Consultant
SalesforceCertified Business Analyst
SalesforceCertified Administrator
SalesforceCertified Platform Developer I
SalesforceCertified AI Associate
SalesforceCertified Platform Foundations
Amazon Web ServicesCertified Cloud Practitioner
2022 – 2023M.S. Business Analytics, The University of Texas at Dallas
2016 – 2019B.E. Electronics & Communications Engineering, Dwarkadas J. Sanghvi College of Engineering
2013 – 2016Diploma in Industrial Electronics, Bhausaheb Vartak Polytechnic

Recognition

  • Insta Award ×2Infosys · performance recognition on the Daimler AG programme
  • Graduate Teaching AssistantThe University of Texas at Dallas · 2023
  • Microsoft Student Ambassador2022 – 2023

Verify: Salesforce Trailblazer profile · AWS on Credly

07Contact

If you need someone who writes the spec and can defend the architecture.

I read everything that comes in. Fastest replies go to specific questions: a role, a project, an org that needs an honest look.